[utils] Transliterate "þ" as "th" (ytdl-org#20897)

Despite visual similarity "þ" is unrelated to "p". It is normally transliterated as "th": $ echo þ-Þ | iconv -t ASCII//TRANSLIT th-TH
bigluke26 · May 10, 2019 · fd35d8c · fd35d8c
1 parent 4eec112
commit fd35d8c
Show file tree

Hide file tree

Showing 2 changed files with 3 additions and 3 deletions.
diff --git a/test/test_utils.py b/test/test_utils.py
@@ -183,7 +183,7 @@ def test_sanitize_filename_restricted(self):
 
         self.assertEqual(sanitize_filename(
             'ÂÃÄÀÁÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖŐØŒÙÚÛÜŰÝÞßàáâãäåæçèéêëìíîïðñòóôõöőøœùúûüűýþÿ', restricted=True),
-            'AAAAAAAECEEEEIIIIDNOOOOOOOOEUUUUUYPssaaaaaaaeceeeeiiiionooooooooeuuuuuypy')
+            'AAAAAAAECEEEEIIIIDNOOOOOOOOEUUUUUYTHssaaaaaaaeceeeeiiiionooooooooeuuuuuythy')
 
     def test_sanitize_ids(self):
         self.assertEqual(sanitize_filename('_n_cd26wFpw', is_id=True), '_n_cd26wFpw')

diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py
@@ -125,8 +125,8 @@ def register_socks_protocols():
 
 # needed for sanitizing filenames in restricted mode
 ACCENT_CHARS = dict(zip('ÂÃÄÀÁÅÆÇÈÉÊËÌÍÎÏÐÑÒÓÔÕÖŐØŒÙÚÛÜŰÝÞßàáâãäåæçèéêëìíîïðñòóôõöőøœùúûüűýþÿ',
-                        itertools.chain('AAAAAA', ['AE'], 'CEEEEIIIIDNOOOOOOO', ['OE'], 'UUUUUYP', ['ss'],
-                                        'aaaaaa', ['ae'], 'ceeeeiiiionooooooo', ['oe'], 'uuuuuypy')))
+                        itertools.chain('AAAAAA', ['AE'], 'CEEEEIIIIDNOOOOOOO', ['OE'], 'UUUUUY', ['TH', 'ss'],
+                                        'aaaaaa', ['ae'], 'ceeeeiiiionooooooo', ['oe'], 'uuuuuy', ['th'], 'y')))
 
 DATE_FORMATS = (
     '%d %B %Y',